Enhancing the Security of Deep Learning Steganography via Adversarial Examples
Steganography is a collection of techniques for concealing the existence of information by embedding it within a cover. With the development of deep learning, some novel steganography methods have appeared based on the autoencoder or generative adversarial networks. While the deep learning based ste...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-08-01
|
Series: | Mathematics |
Subjects: | |
Online Access: | https://www.mdpi.com/2227-7390/8/9/1446 |
_version_ | 1797555161236766720 |
---|---|
author | Yueyun Shang Shunzhi Jiang Dengpan Ye Jiaqing Huang |
author_facet | Yueyun Shang Shunzhi Jiang Dengpan Ye Jiaqing Huang |
author_sort | Yueyun Shang |
collection | DOAJ |
description | Steganography is a collection of techniques for concealing the existence of information by embedding it within a cover. With the development of deep learning, some novel steganography methods have appeared based on the autoencoder or generative adversarial networks. While the deep learning based steganography methods have the advantages of automatic generation and capacity, the security of the algorithm needs to improve. In this paper, we take advantage of the linear behavior of deep learning networks in higher space and propose a novel steganography scheme which enhances the security by adversarial example. The system is trained with different training settings on two datasets. The experiment results show that the proposed scheme could escape from deep learning steganalyzer detection. Besides, the produced stego could extract secret image with less distortion. |
first_indexed | 2024-03-10T16:43:32Z |
format | Article |
id | doaj.art-0bf01d8508234cbfae7d2d4f0cd23074 |
institution | Directory Open Access Journal |
issn | 2227-7390 |
language | English |
last_indexed | 2024-03-10T16:43:32Z |
publishDate | 2020-08-01 |
publisher | MDPI AG |
record_format | Article |
series | Mathematics |
spelling | doaj.art-0bf01d8508234cbfae7d2d4f0cd230742023-11-20T11:46:18ZengMDPI AGMathematics2227-73902020-08-0189144610.3390/math8091446Enhancing the Security of Deep Learning Steganography via Adversarial ExamplesYueyun Shang0Shunzhi Jiang1Dengpan Ye2Jiaqing Huang3Key Laboratory of Aerospace Information Security and Trusted Computing, Ministry of Education, School of Cyber Science and Engineering, Wuhan University, Wuhan 430000, ChinaKey Laboratory of Aerospace Information Security and Trusted Computing, Ministry of Education, School of Cyber Science and Engineering, Wuhan University, Wuhan 430000, ChinaKey Laboratory of Aerospace Information Security and Trusted Computing, Ministry of Education, School of Cyber Science and Engineering, Wuhan University, Wuhan 430000, ChinaKey Laboratory of Aerospace Information Security and Trusted Computing, Ministry of Education, School of Cyber Science and Engineering, Wuhan University, Wuhan 430000, ChinaSteganography is a collection of techniques for concealing the existence of information by embedding it within a cover. With the development of deep learning, some novel steganography methods have appeared based on the autoencoder or generative adversarial networks. While the deep learning based steganography methods have the advantages of automatic generation and capacity, the security of the algorithm needs to improve. In this paper, we take advantage of the linear behavior of deep learning networks in higher space and propose a novel steganography scheme which enhances the security by adversarial example. The system is trained with different training settings on two datasets. The experiment results show that the proposed scheme could escape from deep learning steganalyzer detection. Besides, the produced stego could extract secret image with less distortion.https://www.mdpi.com/2227-7390/8/9/1446steganographyinformation hidingdeep learninggenerative adversarial networksadversarial examples |
spellingShingle | Yueyun Shang Shunzhi Jiang Dengpan Ye Jiaqing Huang Enhancing the Security of Deep Learning Steganography via Adversarial Examples Mathematics steganography information hiding deep learning generative adversarial networks adversarial examples |
title | Enhancing the Security of Deep Learning Steganography via Adversarial Examples |
title_full | Enhancing the Security of Deep Learning Steganography via Adversarial Examples |
title_fullStr | Enhancing the Security of Deep Learning Steganography via Adversarial Examples |
title_full_unstemmed | Enhancing the Security of Deep Learning Steganography via Adversarial Examples |
title_short | Enhancing the Security of Deep Learning Steganography via Adversarial Examples |
title_sort | enhancing the security of deep learning steganography via adversarial examples |
topic | steganography information hiding deep learning generative adversarial networks adversarial examples |
url | https://www.mdpi.com/2227-7390/8/9/1446 |
work_keys_str_mv | AT yueyunshang enhancingthesecurityofdeeplearningsteganographyviaadversarialexamples AT shunzhijiang enhancingthesecurityofdeeplearningsteganographyviaadversarialexamples AT dengpanye enhancingthesecurityofdeeplearningsteganographyviaadversarialexamples AT jiaqinghuang enhancingthesecurityofdeeplearningsteganographyviaadversarialexamples |